The ongoing purpose of this project is to compare the genome structure of pathogenic and nonpathogenic strains of Aleutian mink disease parvovirus (ADV) and to relate these differences to the ability of the various isolates of virus to replicate in cell culture and to cause disease in infected mink. By constructing clones that are chimeric between the nonpathogenic ADV-G (type 1 ADV) and ADV-Utah 1 (type 2 ADV), we have now identified a segment that influences the pathogenicity of ADV. Molecular clones of ADV-G that contained the 73-88 map unit piece of ADV-Utah 2 replicated 2log10 less efficiently than the ADV-G parent in cell culture, but induced antibody and caused disease in mink injected with virus derived from these clones. This segment maps to the capsid protein coding sequences and there are 4 coding differences between the type 1 and type 2 virus. Furthermore, the region is distinct from the 54-63 map unit capsid gene segment that governs permissivity in cell culture. However, computer analysis and comparison with the sequence of canine parvovirus (CPV) suggests that both regions align with ones on the CPV capsid regulating host range and pathogenicity. In order to relate these findings to the ADV virion structure, we have begun work to crystallize ADV. We are utilizing a recombinant baculovirus that synthesizes capsid proteins. The capsid proteins self-assemble into particles with physical characteristics of empty ADV virions. In other work, we have begun studies to analyze structure-function interactions of the ADV P36 promoter. The sequence and spacing of ADV P36 elements differs significantly from other prototypic parvoviruses and these variations may explain the low constitutive and trans-activated strength of the promoter. A number of alterations were produced in the promoter, some of which "restored" promoter strength in CAT assays, but when the changes were introduced into full-length clones of ADV, all the clones were replication defective.